Re: Old Languages w/ new thread

At 8:13 AM -0400 10/04/01, Muke Tever wrote:
>From: "Jesse Bangs" <jaspax@...>
>>In the meantime, I have an unrelated
>>question that's been bugging me--are there any languages or language
>>families unrelated to the Semitic languages that have tri- or
>>bi-consonantal root structure?  How do their structures and forms compare
>>to the Semitic languages?
>
>Indo-European languages have something like it... with its mainly
>biconsonantal roots and vowel change + extensions.
>
>Er, so I gather.
>CoC-o- noun
>CeC- verb,
>CeC-o- adjective
>CoC-eyo- causative verb
>CoC-mo- resultative noun
>etc.
>
>(These may or may not be necessarily true.)
This is all true, but I think it's different from the Semitic type in
an important way. While some of the ablaut patterns in IE have become
grammaticalized, its origin was an accent shift, where *o represents
a vowel from which accent has been retracted. I don't think that
anyone has proposed that the Semitic patterns have an underlying
basis in accent.

That said, I don't think that the Semitic patterns are best described
by the intercalation of vowels and consonants. A particularly
striking example is the Arabic broken plural. The vowel quality of
the plural is only partially determined by the grammatical category,
but all of the plurals begin with a light open syllable followed by a
syllable containing a long vowel: CVCVV{C|CV(V)C}.

Likewise, the shapes of the Hebrew binyanim are also prosodically
determined. In the QAL, the vowel qualities may differ (a-a, a-o,
a-e) but the prosodic shape is the same across categories: .CV.CVC.
(at least for the perfective. (It's interesting to note that the
imperfective of the NIPHAL has the same prosodic shape.)

I just read an article by Robert Hoberman about the Aramaic dialect
spoken by Iraqi Jews living in Amadiya. The binyan system has been
whittled down considerably until there are only two binyanim left:
Binyan I which is monosyllabic, and Binyan II which is disyllabic
(Mark Aronoff gives some interesting historical hints about how this
happened in his book _Morphology by Itself_). Both binyanim
accomodate bi- and triliteral roots. Again, it is the prosodic shape
rather than the intercalation of segments which is the basis for the
distinction between binyanim.

The Yokuts language also has prosodic stem alternations which are
determined by inflectional categories. Verbs can have an underlying
stem prosody -- the available templates are a simple syllable (either
light or heavy), a heavy syllable (the so-called moraic trochee), and
a light syllable followed by a heavy syllable. In addition, verb
suffixes may select a particular stem prosody for the verb, in effect
overriding the default prosody of the verb. It's similar to Semitic
in the interrelationship between prosody and inflectional categories,
but the prosodic shapes are different.

If you're interested in the Yokuts patterns, look at Stanley Newman's
grammar of the language (1944) and Diana Archangeli's various
treatments of the Yawelmani dialect (particularly two papers: one
from 1983 in Natural Language and Linguistic Theory and the other
from 1991, also in NLLT). Be warned: the Newman and Archangeli stuff
is rather technical (in both American Structuralist and Generative
terms) and densely written. But it rewards a careful reading.

Dirk
--